TECRA A9-S9017 -- Idles too hot -- Hardware Support

[Wes Morgan]

On Sun, 31 Aug 2008, freebsd_user at guice.ath.cx wrote:

> In addition to the above URL, other findings:
>
> hw.acpi.toshiba.cpu_speed=7
> hw.acpi.toshiba.force_fan=1
> hw.acpi.thermal.min_runtime=30
> hw.acpi.thermal.polling_rate: 10
> hw.acpi.thermal.user_override=1
> hw.acpi.thermal.tz0.temperature: 61.0C  # On 6.3-RELEASE -p3 -
> 				 	# this value is bogus
>
> hw.acpi.thermal.tz0.active: -1          # Can not change
> hw.acpi.thermal.tz0.passive_cooling: 0  # Can not change
> hw.acpi.thermal.tz0.thermal_flags: 0
> hw.acpi.thermal.tz0._PSV=10.0C          # passive_cooling will-
> 					# activate @ this temp.
>                                        # Above, passive_cooling
> 					# currently N/A.

Have you tried 7.0 instead of 6.3? I can't think of any compelling reason 
to stick with 6.x on a laptop.

> With that being said, please allow me this one (1) rhetorical
> question; how is a common end-user supposed to keep his machine
> running and maintain her/his sanity chasing after issues such as
> we've discussed here?

When dealing with a piece of hardware that was designed from the ground up 
with only Microsoft Windows XP or even Vista in mind, I would say that 
we're doing pretty well.

That the two frequencies for each core disagree, could be simply rounding 
difference or the core may actually report differently, as you may notice 
that your true frequency is not 2200mhz. The ACPI temperature may be 
measured in a different location than the "coretemp". On my system there 
is not even a sysctl for the frequency of the second core. I imagine that 
setting each core to a different frequency could be advantageous if the 
scheduler was able to take advantage of this, but I wonder if some 
applications might be adversely effected.

I would try installing 7-stable and rerunning the tests, perhaps posting 
the results with slightly less verbosity as each message is becoming 
inordinately long and cluttered with too much information.